THE QUINCE 253 



It will be remembered that the hybrids thus 

 produced were of extraordinary growth, but that 

 they produced very few nuts, and that among the 

 seedlings of the second generation there were 

 many trees of dwarfed growth, suggesting the 

 quince hybrids. 



We found reason to believe that the curious 

 result of hybridizing the walnuts might be ex- 

 plained on the supposition that the parent forms 

 had diverged almost to the point of mutual an- 

 tagonism. They had not varied quite to the 

 point where their offspring were sterile, but they 

 were approaching that limit. 



The quinces of the experiment now under con- 

 sideration had diverged one stage farther. They 

 are still within the limits of affinity that permit 

 cross-fertilization and intergrafting, but not 

 within those that permit the production of fertile 

 offspring. Their case is rather to be likened to 

 that of our petunia and tobacco hybrids, which, 

 as the reader will recollect, were lacking in 

 virility and produced no blossoms. The similar 

 case of the motley hybrids made by crossing vari- 

 ous members of the rose family with their cousin 

 the dewberry will be recalled. Also the strange 

 progeny of the strawberry and the raspberry. 



The Chinese-European hybrid quince, then, in 

 its dwarfed growth and its sterility merely illus- 



